WO1985000459A1

WO1985000459A1 - Method and apparatus to improve the positioning accuracy of a tracking arm

Info

Publication number: WO1985000459A1
Application number: PCT/US1984/001019
Authority: WO
Inventors: Andrew M. Rose
Original assignee: Memorex Corporation
Priority date: 1983-06-30
Filing date: 1984-06-29
Publication date: 1985-01-31
Also published as: EP0130836A2; JPS60501782A; IE841686L; AU3154084A; IE55808B1; EP0130836B1; US4556921A; EP0130836A3; DE3480670D1; CA1246736A

Abstract

An apparatus for improving the positioning accuracy of a tracking arm in a disc system and eliminating run-out. The apparatus has acceleration apparatus (20) for moving the arm, braking apparatus (84) to brake the movement of the arm, apparatus for detecting run-out (96), and responsive apparatus (88) for removing the braking current in the event run-out is detected.

Description

METHOD AND APPARATUS TO IMPROVE THE POSITIONING ACCURACY OF A TRACKING ARM

BACKGROUND OF THE INVENTION This invention relates to a method and apparatus to improve the positioning accuracy of a tracking arm as it follows a selected position on the surface of a rotating member, and more particularly, this invention relates to a method and apparatus for use wherein the rotating member is subject to the phenomenon of run-out.

In data storage apparatus of the type using rotating disc or a stack of rotating discs as the recording medium, data is usually recorded in a plurality of concentric data tracks on the surface of the disc. The tracks are closely spaced, for example, at 480 tracks per inch, and the data recorded at densities in excess of 9,000 bits per inch. In order to access the recorded data, it is necessary to move a record and playback head (or transducer) to the selected track in which the desired information is stored, and maintain the head precisely over the center of the track with a minimum displacement error in the presence of any possible displacement error for the whole time the information is being read or being written. As track density increases, it becomes more imperative that the record and playback head, or transducer, be accurately maintained over the selected track in which the desired information is either being read or being written. One factor limiting the accuracy of the tracking position of the transducer over the selected position is the mechanical tolerance in the head positioning motor and the support system. In addition, in some recording discs, which are removable, the abutting surfaces of the disc and the disc drive spindle may cause additional errors. One resultant error from the foregoing mechanical tolerances is the phenomenon of run-out. Run-out is caused by eccentricity of the disc or of the spindle or both. Run-out is the result of the eccentricity or the wobbling of the disc, wherein the center of the disc rotation does not coincide with the center of the data track or of the written material. Run-out may be observed even in common day experience by watching the behavior of a tracking arm in a phonograph record player wherein the arm moves to and fro in a sideways motion as it follows the tracks of a warped phonograph. One technique taught by the prior art of eliminating run-out is disclosed in U.S. Patent No. 3,808,486. Since the disc spins at a constant frequency, run-out occurs at the same known constant frequency. In that patent, the amplifier supplying the power to the actuator which moves the tracking arm has a non- linear response, whereby at the run-out frequency, there is an increased amount of gain from the amplifier to the actuator which moves the transducer. In this manner, the patent teaches that the actuator would be moved an additional amount at the run-out frequency in order to be in phase with the run-out to keep accurate tracking over the tracks. Although this is a viable solution to the problem of run-out, the solution taught by the patent is not desirable for a number of reasons. First, the response time of this servo system is slow. Because both the frequency and the gain have a distribution of values, or a sigma, the increase in gain at the run-out frequency is a broad band. Thus the response factor is slow which requires several revolutions of the disc for the gain increase of this technique to build up. Next, since the teaching of that patent is to change the frequency and gain response of the power amplifier that supplies the current to the actuator, the electronic circuitry to implement that teaching is necessarily complicated and is expensive.

SUMMARY OF THE INVENTION Therefore, in accordance with the present invention whereby a fast response to the problem of run-out is achieved in order to improve the positioning accuracy of a tracking arm to follow a selected position on the surface of a rotating member, the apparatus has acceleration means for moving the arm and braking means for braking the movement of the arm. The apparatus further has means for detecting the run-out and means responsive to the detecting means for removing the braking means in the event of the run-out detected.

The present invention also comprises a method of using the foregoing apparatus. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1(a-d) are various diagrams showing the phenomenon of run-out: (a & b) the impact on the actuator or tracking arm; (c) the plot of distance error versus time as a result of run-out; (d) and a portion of the plot shown in Figure 1(c). Figure 2 is a schematic block diagram of the apparatus of the present invention.

Figure 3 is a circuit diagram of one embodiment of the apparatus of the present invention. Figure 4(a-b) are circuit diagrams of one element of the apparatus of the present invention shown in Figure 3.

DETAILED DESCRIPTION OF THE DRAWINGS Referring to Figure 1(a) there is shown a top view of a magnetic disc 10 typically used in disc drives. The magnetic disc 10 has data tracks recorded on it. One such track 12 is shown for illustration purposes. The disc 10 also has sectors 14 wherein servo information is recorded therein. It should be understood that the sectors 14 and the data track 12 are not physically apparent to the naked eye but are shown in Figure 1 (a) solely for illustrative purposes. The tracks 12 are typically concentric and are recorded such that the center of the circle of the tracks 12 coincide with the hole 16, which is the center of rotation of the disc 10. The phenomena of "run-out" occurs when the disc 10 is mounted on the disc drive as shown in Figure 1(b) and the center of rotation of the disc drive with the disc 10 thereon is not at the hole 16. In fact, for illustration purposes again, the center of rotation is at 16' and one concentric circle around the center rotation 16' is shown by the dashed circle 12'. Thus, the phenomena of run-out, wobbling or precession occurs when a tracking arm 20 attempts to follow a physical concentric circle 12' which does not coincide with the concentric circle 12 in which data is recorded. The tracking arm 20 if allowed to maintain at a fixed distance from the center rotation 16' would follow the concentric circle 12' instead of the track 12 with the data therein. The error induced by maintaining the tracking arm 20 at a fixed distance from the center of rotation 16' is shown in Figure 1(c). For every revolution of the disc 10, the tracking arm 20 would be accurately in place for only two instances of time during the rotation of the disc 10. At all other times, the tracking arm 20 would have an error in position as shown by the plot of distance error versus time in Figure 1(c). The actual tracking of the disc 10 by the arm 20 is shown by graph 22. At every location wherein the graph 22 jumps into the curve 21, (such as points 24a, 24b, and 24c), those points denote when the actuator 20 is over the sectors 14 containing the servo information. At those locations the actuator 20 would have the exact actual positioning information from the reading of the servo information recorded in the sectors 14. When the actuator 20 senses the sector information 14, the information is relayed to the servo feedback circuit (not shown) of the disc drive. Such a circuit is well known in the art. The circuit sensing that the actuator 20 is not correctly positioned would send a current to accelerate or to move the actuator 20 in a direction to reduce the error in distance between the location of the actuator 20 and the actual location of the data track 12. This is shown as curve 26. However, a high bandwidth, fast response servo would automatically reverse the current to the actuator 20, inducing a braking current to slow down the actuator 20 so as not to "overshoot" the intended position, as the actuator 20 nears another servo sector 14. This is shown as curve 28. Such application of the braking current 28 is automatically present in the servo circuit of the prior art. Lower the servo bandwidth would reduce the braking, but would by nature increase the response time and therefore increase the error.

The apparatus of the present invention is intended to improve the positioning accuracy of the actuator 20 as it attempts to follow the data track 12 on the disc 10, wherein the disc 10 has servo information recorded in sectors, and is subject to run-out condition. Furthermore, the actuator 20 is subject to acceleration for moving it and braking means for braking its movement.

Referring to Figure 2 there is shown a block diagram form of the apparatus 50 of the present invention. The apparatus 50 comprises a first source 52 for providing servo information from a first sector 14. A second source 54 provides servo information from a second sector 14. The servo information from the first source 52 and from the second source 54 are entered into a comparator 56. The output of the comparator 56 is entered into a logic means 58. The polarity of the electrical signal from the servo information from the first source 52 or the second source 54 is also entered into the logic means 58. The logic means 58,. based upon the results of the comparator 56 and the polarity of the signal from the first or second servo source 52 and 54 respectively, determines whether or not to activate the switch 62. When the switch 62 is activated, it removes the current to the actuator 20. The timing of the removal of the current to the actuator 20 is provided by the clock 66, such that when- the switch 62 is activated it occurs when the current is a braking current that is applied to the actuator 20. Thus, in Figure 1(d), the apparatus 50 of the present invention would remove the portion of the graph shown as 28. By removing the braking current to the actuator 20, the apparatus 50 of the present invention serves further to allow the actuator 20 to coast in the direction to reduce the distance error between the actuator 20 and the track 12. Thus, positioning accuracy is increased. Referring to Figure 3 there is shown in detail one embodiment,of the apparatus 50 of the present invention shown in circuit diagram form. In the apparatus 50 shown in Figure 3, the servo information 74 is shown as combined from the sector 14 and from a dedicated disc surface. The use of the combination of servo information from the sector 14 and from the dedicated data disc is disclosed in U.S. Patent No. 4,072,990. That patent is incorporated by reference herein. Although in the example shown in Figure 3 the servo information 74 is derived from both the sectors 14 as well as a dedicated disc surface, the apparatus 50 of the present invention may be used with servo information from any source. The input of the servo information 74 to the apparatus 50 is not limited to the information from sector 14 or to the information from the sector 14 in combination with a dedicated disc surface. The servo information 74, from whatever source, is entered into a first operational amplifier 76. The first operational amplifier 76 serves to isolate the source 74 from the rest of the system. The servo information 74 is entered into an R₃ C₃ network along with a first comparator 80 which is then entered into a first exclusive OR gate 96. The servo information 74 is also entered into an R₁ C₁ and R₂ C₂ circuit with a second comparator 78, the output from which is entered into the first exclusive OR gate 96. The network comprising the R₁ C₁ and R₂ C₂ with the second comparator 78 serves as a delay circuit. Thus at the first OR gate 96, the input signals therein are the servo information 74 and a delayed servo information 74. In other words, the first exclusive OR gate 96 compares servo informationfrom two different instances of time. Since the disc 10 is rotating, this is in effect comparing the servo information derived from two positions on the disc 10. The output of the first OR gate 96 is entered into a latch 82. The latch 82 is merely a flip-flop. A clock is gated into the flip-flop 82 and serves to strobe the output of the first exclusive OR gate 96 into the latch 82 at the desired time. From the output of the first comparator 80, the polarity of the electrical signal from the servo information 74 is entered into a second exclusive OR gate 84. The output of the second exclusive OR gate 84 is entered into an AND gate 86. Another input to the AND gate 86 is from the

output of the latch 82. From the AND gate 86 the output is connected to the switch 88 which removes the power to the amplifier which controls the current to the actuator 20. In the configuration shown in Figure 3, the switch 88 is switched off.

The theoretical basis for the apparatus 50 of the present invention may be understood as follows. The apparatus 50 analyzes the servo information from two locations on the disc 10. This may be done, for example, by comparing two servo information that differ in time, which is that shown and described in Figure 3. The effect of delay in time on a moving body is of course a change in distance. Thus, the apparatus 50 as shown in Figure 3 compares the servo information from two different physical locations on the disc 10. If the comparison between one location to the other location shows a difference, then this indicates that either the actuator has moved or the disc 10 has moved with respect to the actuator. The result of the comparison from the first OR gate 96 is to set the latch 82 to indicate that potentially the condition of run-out has occurred. If the later produced signal from the servo information 74 shows the error from the track 12 is smaller than the earlier produced signal from the servo information 74, then this indicates that the actuator 20 is getting closer to the actual data track 12. The first OR gate 96 compares the later produced electrical signal from the first comparator 80 to the earlier produced signal from the second comparator 78. If run-out has occurred, the result of the first OR gate 96 would be a negative signal entering into the latch 82. A clock serves to clock in the signal from the first OR gate 96 after the information from the sector 14 has been read in. If the result of the first OR gate

96 is a negative signal entering into the latch 82, the Q output of latch 82 would of course be a positive signal which is entered into the AND gate 86. If at this point the braking current is applied then the electrical signal from the output of the second exclusive OR gate 84 is positive going, which is entered into the other input to the AND gate 86. The AND gate 86 with two positive input pulses thereinto would produce a positive pulse activating the switch 88 which would cut off the current to the power amplifier to produce the current to the actuator 20. This serves to remove the braking current to the actuator 20.

In an improvement over the apparatus 50 of the present invention, a compensator 90 is provided between the servo information 74 and the output of the apparatus 50 to the power amplifier. The compensator 90 is shown schematically in two different embodiments in Figures 4(a & b). The effect of the compensator 90 is to impart minimum acceleration current to the actuator 20. This is graphically demonstrated in Figure 5, wherein the acceleration current is shown as curve 92. When the acceleration current decreases and reaches a minimum point as denoted by the level entitled "Min", the acceleration current continues at that minimum level rather than taking on the dashed curve of 92'. When the time to accelerate has ended, the acceleration current drops from the .minimum level down to zero. The braking current as shown by curve 94 is of course removed by the apparatus 50 of the present invention. The use of a minimum level acceleration current in the apparatus 50 of the present invention increases the positioning accuracy of the actuator 20. Referring now to Figures 4 (a & b), these are well known compensators of conventional design. In Figure 4a, the compensator 90 is a lead/lag phase compensator. The position signal is entered into the operational amplifier 100 through a R₇C₇R₈ network. The R₇C₇R₈ network serves to change the position signa to one of rate of change. As the position signal decreases, indicating a reduction in error, the rate of change becomes negative. The other input to the operational amplifier 100 is from the R₁₀R₁₁ network, which is a constant. Once the value of R₁₀R₁₁ exceeds the output of R₇C₇R₈, the output of the operational amplifier changes polarity and current to the power amplifier is maintained at a constant level corresponding to the level "Min" shown in Figure 5. In Figure 4b, the compensator 90 is a tachometer type compensator. Again, the compensator is of well known conventional design.

There are many advantages to the method and apparatus of the present invention. Unlike the apparatus taught by U.S. Patent No. 3,808,486, the method and apparatus of the present invention is fast in response. By removing only the counter-productive braking current, the loop gain and stability of the system is maintained. Finally, by imparting minimum acceleration current, the additional compensation circuit increases the positioning accuracy of the servo network.

Claims

WHAT IS CLAIMED IS:

1. An apparatus to improve the positioning accuracy of an actuator adapted to follow a selected track on a magnetic disc which has servo information imbedded in sectors of said disc, wherein said disc is subject to run-out and said actuator is subject to acceleration current and braking current as said actuator follows said track, said apparatus comprising: means for detecting said run-out; and responsive means for blocking said braking current in the event of said run-out detected.

2. The apparatus of claim 1 wherein said detecting means comprises: means for generating a first electrical signal based upon servo information from one of the sectors of the surface of said disc, containing servo information; means for generating a second electrical signal based upon servo information from another of the sectors of the surface of said disc, containing servo information; and means for comparing said first and second signals.

3. The apparatus of claim 2 wherein said responsive means comprises: switching means for removing said braking current; and logic means responsive to said comparing means and polarity of either first or second signal for applying said switching means.

4. The apparatus of claim 3 further comprising minimum acceleration means for imparting minimum acceleration current.

5. A method of improving the positioning accuracy of an actuator in following a selected position on the surface of a rotating member, wherein said member may be subject to run-out, said apparatus having acceleration means to move said actuator and braking means to brake said movement, the method comprising detecting said run-out; and removing said braking means in the event of said run-out detected.

6. The method of claim 5 further comprising imparting minimum acceleration means to said actuator.